Bayesian Non-Homogeneous Hidden Markov Models for Time-Varying Functional Connectivity

• Activity Number: 204 - Bayesian Methods for the Analysis of Complex Brain Imaging Data
• Type: Topic-Contributed
• Date/Time: Tuesday, August 10, 2021 : 1:30 PM to 3:20 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Jaylen Lee* and Michele Guindani and Marina Vannucci and Ryan Warnick
• Companies: University of California, Irvine and University of California Irvine and Rice and Rice University
• Keywords: Dynamic; Bayesian; HMM; fMRI; Brain; Connectivity

Abstract

Time Varying Functional Connectivity (TVFC) aims at investigating how the brain’s functional networks evolve through the course of an fMRI experiment. We discuss a multivariate Bayesian framework where the networks are estimated through the classification of latent neurological states and thereby leads to the estimation of sparse connectivity networks in an integrated framework that borrows strength over the entire time course of the experiment. We consider the use of Hidden Markov Models for the automatic learning of the states and the associated transitions. We further discuss how the assumption of stationary transition kernels throughout the experiment may not be appropriate to fully capture the changing responses of a subject during the experiment. We thus propose a Bayesian model with time-varying transitions informed by exogenous subject- and experiment- specific variables. We apply our modeling framework to the analysis of fMRI squeezing task data.